green1
Active Member
Which is a patently ridiculous claim, they simply don't have enough control of all the factors to do that.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
TACC failed to brake at stop, nearly accident.
- Thread starter Lintrix
- Start date
-
- Tags
- Driving Dynamics Model S
Sure they do, they just build them without wheels or doors, and bury them behind the factory.
If you have sensors on each side view mirror, you could see the further car around the nearer car. If it is smart, it can also look in reflections (shop windows, other vehicle surfaces, etc) and compare colors and shapes to what it knows already from what it saw in front of it, and also looking through windows of vehicles ahead. This is what humans do anyway (who are decent drivers), so it would basically be mimicking what humans do, only on a more mechanical scale. A lot of programmers get lazy and claim this is "impossible", but clearly it is not only possible but within the realm of a corporation to easily do with many (properly organized) engineers.
jgs
Active Member
Larry Niven wrote about the solution in "Safe At Any Speed", but he doesn't project availability until 3100 AD. You can get on the waiting list now for a $5000 nonrefundable deposit, though. Safe At Any Speed by Larry Niven — Reviews, Discussion, Bookclubs, Lists
Yes, that obviously would work great. Also just directly looking at tires ahead from underneath would work very well (straight, cross angles, etc.). I hadn't thought of those. Those are great.
- - - Updated - - -
That seems like an insufficient warning and poor instructions. From what other people explained here, if you are in the #1 lane on I-280 south at 4PM workday coming up to the stoppage area before the carpool lane starts somewhere before CA-85, if traffic is heavy, the car in front of you can be at a dead stop not too far around a gentle curve, and you are using "TACC in highway use" and you will slam into the car in front of you at 79MPH. So you used it as Tesla lawyers described, but it failed.
The explanation by rneugebauer in post #24 TACC failed to brake at stop, nearly accident. - Page 3 is much clearer, understandable, precise, logical, and useful to a driver, and seems more accurate by anecdote, than what the lawyers wrote. This is a case of I think the lawyers are culpable for killing people and property damage.
*Exactly* how my 2007 Infiniti M35 behaved. Left-hand exits on the highway were extremely annoying (when there were slow cars in the right lane and I was passing... of course I'd normally be in the right lane for normal travel ;-) ).
Luckily, I have a pre-Oct 2014 car so I don't need to worry about whether TACC / AEB etc are engaging or not
I don't see the difference between surface streets and highways.
When I think highway, I think highway CA-17, I-280, US-101, I-880, I-680, and further east, CA-152, I-580, I-5, CA-120, CA-99. These are often jammed packed, often fast moving, often many lanes, often people coming and going, and some of them have cross traffic and construction crews moving about in odd directions. CA-152 has stoplights and freeway portions. CA-132 is the same. You can be driving down a large freeway then suddenly run into a meter light for a transfer to another freeway, solid with traffic creeping or stopped. Farm areas are even more interesting in dissolving this distinction.
Many surface streets have long expanses which allow fast driving without a lot of cross traffic, and then you get to areas where there is cross traffic.
Any kind of autopilot would have to know what the driver intends to do. I.e., in the example here, if the Prius turning right was the direction you wanted to go in, then autopilot would follow that vehicle so as to not hit any other vehicle and not hit the Prius. But in the example here, if you actually wanted to go the direction of the stopped light, the autopilot would similarly need to know that you intended to go the direction of the stopped light ("straight" in driving nomenclature). TACC is not able to interpret your intentions and actual future actions, if for no other reason than the programmers don't know whether or not you know whether or not you put your turn signal on or not (or, for that matter, are a good signaler). Obviously, a good autopilot should predict probable outcomes by looking at all sensory input, interpreting movements, signals, visuals, attitudes, etc. But I think this is a key difference between the current infantile version of TACC and what autopilot would be.
Let me say that distinction a bit more: if the autopilot is driving, it knows where it could and will try to go. If the autopilot isn't driving, it (TACC, assist, autopilot, etc.) doesn't know what you (the driver) are intending and are going to decide to do and end up doing. * This is a HUGE difference between what we've all seen Google Driverless cars do and what Tesla Autopilot features are currently available. We are easily confused by watching all the Youtubes Google puts out and thinking Tesla is more modern than that (it isn't **).
I think the programmer to driver communication from Tesla ought to be better. You'd think that the manual writer in Palo Alto software division would just put it on the instruction manual and all the Silicon Valley Tesla customers would read that manual and everything would be easy. But instead I have this foreboding fear that the driver instruction manual is actually written in a foreign language dozens of timezones away by insular software programmers, routed through Southern California, East Coast Lawyerville (Iowa/Connecticut), Palo Alto and Fremont before getting sent to translators in China and India then back to Palo Alto, many months after the actual software is released.
---
* What if you intended to go right, had your turn signal on, and there wasn't space in the lane for your fat Tesla to fit between the stopped car and the right curb even though the skinny Prius easily made it? What if there was room, but you didn't get far right enough because you thought there wasn't? What is driver assist supposed to do then? Autopilot would figure it out (supposedly), but driver assistance would have to be far more superior to that to guess whether or not you were going to try to squeeze through, if it even knew with confidence you were going that way to begin with.
** They say LIDAR units cost $150,000.00 (the things Google driverless cars use).
Last edited:
stopcrazypp
Well-Known Member
I think you are underestimating how hard it is to mimick what humans do. Our brain does a lot of processing behind the scenes of visual information and makes a lot of qualitative decisions about identifying things.
For example, a human can easily distinguish between a human, bicycle, motorcycle, car, debris, etc. in a split second (as well as direction of motion) and in all kinds of lighting conditions, but our cameras + AI aren't able to do so easily. Not to mention the computational load might be too much for a typical car computer to handle. That why a lot of the autonomous vehicles rely mainly on radar or lidar for road conditions and not a camera system (beyond relatively simple things like road signs).
You're welcome.
- - - Updated - - -
I think you pegged it. I wasn't aware I had such prejudices seeping through but there they are.
- - - Updated - - -
I admit I have to defer to your current experience of the state of the art in this. I thought we were already further down this road than we must be.
Krugerrand
Meow
Ok, the autonomous parts of the brain are amazing. What judgments we make with the info that gives us, well....
Krugerrand
Meow
stopcrazypp
Well-Known Member
Even Elon (who is relatively optimistic) says we are still 5-6 years off from fully autonomous. And he doesn't say we can accomplish it using only camera systems (more expensive lidar systems like Google uses might turn out to be necessary).
The issue is coming up with algorithms that can reliability process the images. That's not something putting in long hours of programming or even throwing a lot of computational resources (like a supercomputer) at the problem can solve. At the current state, lidar remains the system that is reliable enough for full autonomous driving (and even then Google is limiting to 25mph use). Computer Vision AI and current camera systems are still not reliable enough without human intervention. They are used in emergency braking systems (which don't guarantee stopping in time to prevent collision, only reducing the impact). Volvo's late-2010 pedestrian detection system famously failed 3 times out of 12 demonstrations, which obviously isn't good enough for autonomous driving.
CHG-ON
Still in love after all these miles
Yes, we must always be on top of it. I have had the same thing. My assumption is that TACC is locked on the Prius, which is still moving, so it is simply trying to follow it. We are WAY too early in this tech to simply let go and allow tech to take care of us. Personally, I think that will take quite some time to get there. Some sort of AI? Just don't know.
Anyone who has used adaptive cruise control on other cars will recognize that this worked exactly as it is supposed to. Nothing to see here folks -- just new customers learning how adaptive cruise control works. It will only stop for a stopped car if it earlier locked onto that car while it was moving. Its pretty simple.
Is this confirmed? I believe I may have had TACC lock onto a stopped vehicle and successfully stopped normally before. Although maybe even the slightest movement could trigger the TACC to lock.
I understand that you guys are comparing TACC to other adaptive cruise control on other brands, but isn't TACC far too much different than other ones to say that they all work the same?
Similar threads
